INDEX

Explanations

names of locations or brands containing the word "fort"

oai_token-act-pair · gpt-3.5-turbo

references to the term "fort" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Magikarp

-0.68

WAYS

-0.67

PER

-0.64

Topics

-0.63

OP

-0.63

MIT

-0.63

 Principle

-0.61

 liquids

-0.61

ONS

-0.60

Cash

-0.59

POSITIVE LOGITS

fort

1.25

uitous

1.21

ieth

1.06

dinand

0.98

itude

0.92

cest

0.91

icio

0.90

ress

0.87

uration

0.87

iors

0.87

Activations Density 0.009%

names of locations or brands containing the word "fort"

references to the term "fort" in various contexts

No Comments

No Known Activations